Bruteforce attack useless

[Revelati]

Hmm, interesting links. Ive been wanting to try and build a jammer for a while now but have been too busy.

Im working on updating the script to launch the transparent version of KMS as we speak. Ive gotten it to work from a fresh boot off bt3 live, however I have run into a few problems.

Basically I have gotten the script to setup everything correctly so that at0 is put up and configured. Dhcp starts as long as you have the proper file in /etc/dhcpd.conf, and the route/iptables seem to be setup correctly too.

What is odd is that there is a huge performance difference client side. If I setup everything manually everything runs very smoothly, pages load quickly and there are few lost packets or disconnects.

If I use the script, clientside performance becomes horrible. I can ping outside ips but about half timeout and the other take 1-2 seconds compared to 75-150ms for the manual configure. DNS requests are also affected, one in 5 makes it out to the net and pages only load text or timeout before images load.

Wireshark is showing massive packet loss between my two interfaces, it is also showing frequent disconnects and spamming dozens of arp requests at frequent intervals.

The perplexing thing is that when I check my configuration outputs they all seem to match up, the route seems to be added correctly, iptables match my manual config, ip forwarding must be on or it wouldnt get through at all. So ive got a bunch of debugging to do I guess till I can find out whats causing this slowdown.

The other thing that isnt working quite right is my cleanup function. routes and tables are added correctly but when I shut down the program they aren't being deleted which means that if you want to stop and restart the script you need to reboot.

Sorry for my noobishness on this, im sort of learning as I go since i've never had any sort of formal training in programming. Ahh well, hope yu guys are still making progress!

[hm2075]

made alot of progress,

transparent and non transparent mode workng
got wireless keys via meterpreter
installed meterpreter as a service

video coming up, uploading as we speak.... as well as all commands,

hope no one has any objections with me starting a new thread,

the commands should give everyone a head start, and if its a new thread users won't have to go through seven pages to find a good starting point

edit : put lots of sleep commands in your scripts, that way it will work, i had the same problem when i tried to script

more edits : If you want wireless key viewer to be undetected by av's, unpack it via upx, hex edit the http: links to its source and you will get rid of the bulk of the signatures, the only one that remains is the pswtool signature which is only detected by f-secure and another av

[Revelati]

Great! can't wait to see it!

My performance problem turned out to be weirder than I thought. When I am on my client computer I can surf google. and a few other pages just fine. Pings to google. come back in under 100ms When I go to yahoo.com however it just hangs indefinately, doesnt load and wireshark gets tons of

ICMP Destination unreachable (fragmentation needed)
TCP [TCP Retransmission] [TCP segment of a reassembled PDU]
HTTP [TCP Retransmission] GET / HTTP/1.1

over and over again.

[terminal86]

But please do it here in the wireless forum, and please post the link dude
Perhaps you could link this thread as a kind of reference, to avoid that the things we already pointed out are discussed over and over again.

Continuative thread:

http://forums.remote-exploit.org/showthread.php?t=18369

[Blackrose]

Nowadays all that theory about how long it would take to process and generate keys has gone out the window with GPU computing.

does anyone know how long it would take to crack a 63 character phrase like the ones GRC distribute with a single new nvidis tesla?

[terminal86]

Elcomsoft claimed its Password Recovery product, which can also be used in a distributed fashion across a network for faster cracks, could speed up WPA/WPA2 passphrase guessing by a factor of 100. But what did that mean in practical terms? How long a passphrase and how fast to crack it?

I got an answer from Elcomsoft's Andrey Belenko, who said via e-mail that without dictionary words being involved, cracking is still quite intensive: perhaps three months to crack a lowercase-only random eight-character password using a PC with two Nvidia GTX 280 video cards. That's on the order of what seems reasonable, given the underlying algorithm's strength. There are 200 billion possible passphrases in this format, and some substantial hashing overhead to turn a passphrase into the WPA/WPA2 key material.

Belenko said that Elcomsoft speeds key recovery rates from 400 passwords per second on a mid-range Core 2 Duo before the GPU acceleration; that moved to 12,500 per second with a single GTX 280, and 50,000 with the Tesla S1070 ($8,000 street). For about $15,000 worth of hardware, Belenko believes that an 8-character lowercase password could be cracked in perhaps a week, a dramatic drop in cost versus recovery time and key length.

http://arstechnica.com/news.ars/post...passwords.html